1. Field of the Invention
The invention relates to metal recovery from iron-bearing ores and somewhat more particularly to a method and apparatus for direct recovery of metal from relatively fine-grained metal-bearing ores and/or concentrates thereof wherein carbon and/or hydrocarbon-containing materials are utilized as the initial reduction materials.
2. Prior Art
The classical recovery of metal, such as iron-steel, in a metal recovery furnace system, such as a blast furnace-steel furnace system requires extensive preparation of metal-bearing ores and/or fine-grained metal-bearing ore concentrates, along with the refining of coal into high-grade metallurgically stable coke.
Lump ores, such as were conventionally utilized as starting materials during the time when the blast furnace technology began, are only today available on a small scale and less than about 50% of this type of material is now being directly utilized. Today, more than about 85% of the first refinements of a metal-bearing ore for metal, such as iron recovery are with fine-grained ore concentrates. However, typical blast furnaces can only be operated with lumpy starting materials and very special properties are required of such starting materials, particularly in regard to their physical structure. For this reason, typical metal ore concentrates, such as iron ore concentrates must first be subjected to a thermal sintering process or to a two-stage mechanical-thermal solidification process.
The fossil reduction medium typically utilized to reduce metal-bearing ores, coal, must be converted, via a separate process in a coke-oven means, into a coke material which has sufficient stability for conditions encountered within a blast furnace. Only special types of coal can be utilized to produce such coke and this coal is available only in everdecreasing quantities.
Accordingly, the art has undertaken numerous attempts to develop methods of metal recovery which are less costly in regard to the preparation of starting materials and which are more compatible with the type of raw materials available today.
One such attempt comprises combining known processes and has become known as the "direct reduction process." In these processes, a series of methods have been proposed wherein coke is no longer utilized for reduction, but instead, a reducing gas is utilized as the reduction agent and/or wherein fine ore particles are no longer sintered with coke dust but are instead utilized in the form of pellets, which are less costly to manufacture. Suitable shaft furnaces and retort means have been developed for use as reduction housings with such processes.
The prior art endeavors to advance a further step and reduce fine-grained metal-bearing ores directly with a gas have led to a plurality of diverse methods, in which ore particles and a gas are brought into contact to react with one another in a suspended state; which, if achievable, would produce a considerable technological and economic advancement. However, attempts to bring these methods into operation maturity has not heretofore been successful. The reason for this appears to lie in the fact that it has not been possible to combine the various diverse technologies in such a meaningful manner that the deficiencies and limits of the various independent steps, which are known per se, can be adequately eliminated.
For example, in regard to the reduction step, a substantial art concentration has occurred on the technological processes utilizing such variants as turbulence layers, airborne dust clouds and/or jet smelting in a direct current system; all of which require a very narrow range of gas-to-solid ratios in order to maintain a stabile system (for example, see Bogdandy and Engell, "The Reduction of Iron Ore" (1967), pages 209-243, published by Stahleisen mgH/Dusseldorf, Germany).
Reduction processes involving the so-called airborne dust cloud as well as those involving the so-called turbulence layers utilize, as a common typical characteristic feature, comparatively low relative speeds between ore particles and reducing gas so that considerable amounts of gas are necessary to attain the heat requirements of the reduction reaction and for ore heating.
In producing an airborne dust cloud, gas speeds are selected which are exactly sufficient to maintain extremely fine ore particles in gas suspension relative to the specific weight of such particles and the falling speed resulting therefrom. Accordingly, gas speeds are regulated so as to be, as a rule, below about 1 m/sec. and in exceptional cases, where the average particle size is over about 0.1 mm, gas speeds above about 1 m/sec. are possible.
In the turbulent layer methods, gas speeds are utilized which are on the same order of magnitude as set forth above. With these methods, it is possible to work with comparatively large particle sizes. In instances where the average particle size is over about 1 mm, the upper limit of the gas speed is on the order of magnitude of about 1.5 m/sec.
In contrast, with the so-called "jet-smelting" process, the direct-current principle has been utilized whereby the influx gas speed and the particle falling speed are actually adapted to one another so that almost no relative speed between the solids and the gas exists.
Reduction of fine size ore particles occurs extraordinarily fast via a heated gas so that the reduction equilibrium of a reducing gas at a specified reaction temperature must be considered. After an extremely short contact period between ore particles and a reduction gas, gas molecules are absorbed onto the immediate or outer surface of an ore particle so that the reduction gas is unable to cause further reduction on such particle. Accordingly, with the aforesaid prior art methods, it is apparent that because of the limiting low gas flow speeds over long periods, extremely large amounts of reducing gas must be provided to the ore particles in order to attain at least a somewhat satisfactory degree of reduction. Naturally, this results in an extremely unsatisfactory utilization of the reduction gas, and is characterized by a high ratio of gaseous to solid materials, which as a rule amounts to more than 2 to 3 times the stoichiometrically required amount of gas. Thus, for example, the J. Iron Steel Inst., Vol. 194 (1960), pages 211-221, indicates that the time required for a 80% reduction of fine-grained ore particles of varying origin pulverized into very small particles (50 to 150 microns) and in a state of suspension within a hydrogen stream and heated to a temperature in the range of 700.degree. to 1100.degree. C., is between 20 to 30 seconds; indeed, almost independent of the type of ore and/or particle size thereof.
Further, it is known that a high degree of chemical utilization of a reduction gas can be attained if the reduction reaction is conducted at as high a temperature as possible. However, experiments have shown that fine-grained ore-mixtures reduced at temperatures above 900.degree. C. undergo, depending on the degree of reduction, a more or less surface-softening reaction so that the use of higher reduction temperatures is not practical. To the contrary, in the known direct reduction processes, suggestions are always made to the effect that the reduction temperature must be kept below this limit, i.e., below a temperature at which the particles undergo a melt-phase formation. However, the chemical and thermal dynamic laws (equilibrium ratio), with a given reduction temperature below 900.degree. C., dictate a comparatively low material conversion, even when a complete utilization of a reduction gas occurs. Accordingly, this condition, in itself, is an even more compelling reason for conveying particularly large quantities of a reduction gas over very extended periods of time past the ore particles in order to repeatedly disrupt the dynamic thermal equilibrium about each particle and effect a good reduction and, in this manner, produce favorable conditions for further reduction. However, this can only be achieved with the above-process techniques if long dwell periods are acceptable, which necessarily cause low material outputs.
In order to produce somewhat more favorable conditions in this regard, certain prior art has suggested reducing fine-grained iron ore in a dry state within a cyclone heat-exchange system via a reducing gas. This process seeks to produce finely-grained iron powder. However, taken by itself, this has a decisive shortcoming in that, given the necessary temperature, there is a very great danger of a reoxidation of the ore (which has already been reduced) occurring on impact with air. To avoid this danger, it has been suggested that this effect can be countered by the addition of coal in the individual cyclone stages. Iron powder produced in this fashion, sometimes referred to as iron sponge, must be hardened or stabilized with a protective gas and then conveyed for further processing. Further, the presence of solid carbon in iron sponge necessarily results in a poor utilization of the reducing agent.
Production of reducing gas for use in iron ore processing, outside the blast furnace technique, is known and a majority of such processes heretofore utilized a starting material comprised of natural gas which was converted into a CO/H.sub.2 -rich reduction gas via a cracking procedure. However, there are also known processes for producing a reduction gas directly from coal. Such processes take place in so-called gas generators, of which only the high temperature-smelting generators directly supply a gas suitable for reduction purposes and without apprecaible amounts of hydrocarbons therein. However, such gas is not free of CO.sub.2, the presence of which is not desirable in a reduction gas.
Recently, a gasification process has been developed wherein coal is blown into a molten iron bath and is thereby converted into a reduction gas which is free of hydrocarbons and CO.sub.2. In this process, coal is conveyed in a fine particle form into a molten iron bath via a water-cooled sparger-like device and simultaneously oxygen is introduced into the molten iron bath via a second sparger-like device. Lime is added to such iron bath in order to produce an alkaline slag on such bath so that any sulfur present in coal is absorbed by the alkaline slag. With this method, cheap and impure coal can be successfully converted into a useful reduction gas having a high CO/H.sub.2 concentration. Further, such reduction gas is typically produced at a temperature of about 1400.degree. C., which is sufficiently high for reduction processes.
However, a problem exists in incorporating the aforesaid conversion of coal into a reduction gas via a metal bath with an metal-ore recovery process so as to achieve a technologically useful, problem-free and economical process which overcomes the performance or product limits of presently known reduction systems.